from sklearn_benchmarks.reporting.hp_match import HpMatchReporting
import pandas as pd
pd.set_option('display.max_colwidth', None)
pd.set_option('display.max_columns', None)
pd.set_option('display.max_rows', None)

scikit-learn vs. scikit-learn-intelex (Intel® oneAPI)

measurements = ['iteration_throughput', 'latency', 'mean_duration_sklearn', 'mean_duration_sklearnex', 'speedup', 'std_duration_sklearn', 'std_duration_sklearnex', 'std_speedup']

def get_position(string):
    if "mean_duration" in string:
        return 3
    elif "std_duration" in string:
        return 2
    elif "score" in string:
        return 1
    elif "speedup" in string:
        return 0
    else:
        return -1

sorted(measurements, key=get_position, reverse=True) 

['mean_duration_sklearn',
 'mean_duration_sklearnex',
 'std_duration_sklearn',
 'std_duration_sklearnex',
 'speedup',
 'std_speedup',
 'iteration_throughput',
 'latency']

reporting = HpMatchReporting(against_lib="sklearnex", config="config.yml", log_scale=True)
reporting.make_report()

We assume here there is a perfect match between the hyperparameters of both librairies. For a given set of parameters and a given dataset, we compute the speedup time scikit-learn / time sklearnex. For instance, a speedup of 2 means that sklearnex is twice as fast as scikit-learn for a given set of parameters and a given dataset.

KNeighborsClassifier_brute_force

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: algorithm=brute.

WARNING! Mismatch between validation scores for 14 prediction (58.33%). The observed differences can be found in the diff_accuracy_scores column. The chosen difference threshold is 0.001 for accuracy_score. See details in the dataframe below.

	estimator	function	n_samples_train	n_samples	n_features	parameters_digest	dataset_digest	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	algorithm	n_jobs	n_neighbors	accuracy_score_sklearn	mean_duration_sklearnex	std_duration_sklearnex	accuracy_score_sklearnex	speedup	std_speedup	diff_accuracy_scores
1	KNeighborsClassifier_brute_force	predict	100000	1000	100	e70d8e7886766d41f30889506baa1e67	2cf49829391aa7891e877f2ed070adf0	1.951474	0.179328	NaN	0.000410	0.001951	brute	-1	1	0.663	0.169336	0.007575	0.687	11.524242	11.535767	0.024
4	KNeighborsClassifier_brute_force	predict	100000	1000	100	22d5eac4a4146149b35eae6914921514	2cf49829391aa7891e877f2ed070adf0	2.721689	0.029252	NaN	0.000294	0.002722	brute	-1	5	0.757	0.170137	0.002389	0.742	15.997088	15.998665	0.015
7	KNeighborsClassifier_brute_force	predict	100000	1000	100	16f94fc93942ff7ece860c9b0f64645f	2cf49829391aa7891e877f2ed070adf0	2.064564	0.003485	NaN	0.000387	0.002065	brute	1	100	0.882	0.205597	0.000490	0.875	10.041787	10.041815	0.007
8	KNeighborsClassifier_brute_force	predict	100000	1	100	16f94fc93942ff7ece860c9b0f64645f	2cf49829391aa7891e877f2ed070adf0	0.022841	0.000331	NaN	0.000035	0.022841	brute	1	100	1.000	0.008809	0.000219	0.000	2.592934	2.593735	1.000
10	KNeighborsClassifier_brute_force	predict	100000	1000	100	2d5f2b2c02d77766434d9e5033b4a76c	2cf49829391aa7891e877f2ed070adf0	2.765083	0.021127	NaN	0.000289	0.002765	brute	-1	100	0.882	0.207966	0.004242	0.875	13.295860	13.298626	0.007
11	KNeighborsClassifier_brute_force	predict	100000	1	100	2d5f2b2c02d77766434d9e5033b4a76c	2cf49829391aa7891e877f2ed070adf0	0.024224	0.003025	NaN	0.000033	0.024224	brute	-1	100	1.000	0.008989	0.000771	0.000	2.694897	2.704790	1.000
13	KNeighborsClassifier_brute_force	predict	100000	1000	100	efac583623711d395ddae7a1e881ff68	2cf49829391aa7891e877f2ed070adf0	2.046789	0.002251	NaN	0.000391	0.002047	brute	1	5	0.757	0.168089	0.000189	0.742	12.176808	12.176816	0.015
16	KNeighborsClassifier_brute_force	predict	100000	1000	100	b79d63bb7670c492de5c3befac58fe29	2cf49829391aa7891e877f2ed070adf0	1.221240	0.041680	NaN	0.000655	0.001221	brute	1	1	0.663	0.167715	0.000769	0.687	7.281648	7.281725	0.024
19	KNeighborsClassifier_brute_force	predict	100000	1000	2	e70d8e7886766d41f30889506baa1e67	88843f54689e3271092f70126e1de585	1.726383	0.025467	NaN	0.000009	0.001726	brute	-1	1	0.896	0.025483	0.000099	0.967	67.745479	67.745988	0.071
22	KNeighborsClassifier_brute_force	predict	100000	1000	2	22d5eac4a4146149b35eae6914921514	88843f54689e3271092f70126e1de585	2.610355	0.024594	NaN	0.000006	0.002610	brute	-1	5	0.922	0.026602	0.000407	0.974	98.126578	98.138068	0.052
25	KNeighborsClassifier_brute_force	predict	100000	1000	2	16f94fc93942ff7ece860c9b0f64645f	88843f54689e3271092f70126e1de585	1.970619	0.002506	NaN	0.000008	0.001971	brute	1	100	0.929	0.059669	0.001612	0.975	33.026029	33.038084	0.046
28	KNeighborsClassifier_brute_force	predict	100000	1000	2	2d5f2b2c02d77766434d9e5033b4a76c	88843f54689e3271092f70126e1de585	2.636970	0.022323	NaN	0.000006	0.002637	brute	-1	100	0.929	0.059686	0.001518	0.975	44.180964	44.195255	0.046
31	KNeighborsClassifier_brute_force	predict	100000	1000	2	efac583623711d395ddae7a1e881ff68	88843f54689e3271092f70126e1de585	1.946543	0.001387	NaN	0.000008	0.001947	brute	1	5	0.922	0.026544	0.000196	0.974	73.331700	73.333696	0.052
34	KNeighborsClassifier_brute_force	predict	100000	1000	2	b79d63bb7670c492de5c3befac58fe29	88843f54689e3271092f70126e1de585	1.082449	0.008280	NaN	0.000015	0.001082	brute	1	1	0.896	0.025576	0.000072	0.967	42.322488	42.322658	0.071

Raw results

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.581	0.0	-1	1	0.050	0.005	0.245	0.247	See	See
3	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.765	0.0	-1	5	0.048	0.001	0.245	0.245	See	See
6	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.596	0.0	1	100	0.048	0.001	0.254	0.254	See	See
9	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.441	0.0	-1	100	0.048	0.001	0.261	0.261	See	See
12	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.602	0.0	1	5	0.048	0.001	0.254	0.254	See	See
15	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.514	0.0	1	1	0.048	0.000	0.257	0.257	See	See
18	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.004	0.0	0.362	0.0	-1	1	0.008	0.000	0.527	0.527	See	See
21	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.004	0.0	0.367	0.0	-1	5	0.008	0.000	0.520	0.520	See	See
24	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.004	0.0	0.374	0.0	1	100	0.008	0.000	0.510	0.510	See	See
27	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.004	0.0	0.364	0.0	-1	100	0.008	0.000	0.524	0.525	See	See
30	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.004	0.0	0.376	0.0	1	5	0.008	0.000	0.506	0.506	See	See
33	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.004	0.0	0.365	0.0	1	1	0.008	0.000	0.522	0.522	See	See

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	KNeighborsClassifier_brute_force	predict	100000	1000	100	1.951	0.179	0.000	0.002	-1	1	0.169	0.008	11.524	11.536	See	See
2	KNeighborsClassifier_brute_force	predict	100000	1	100	0.025	0.002	0.000	0.025	-1	1	0.009	0.000	2.899	2.900	See	See
4	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.722	0.029	0.000	0.003	-1	5	0.170	0.002	15.997	15.999	See	See
5	KNeighborsClassifier_brute_force	predict	100000	1	100	0.025	0.002	0.000	0.025	-1	5	0.009	0.000	2.866	2.866	See	See
7	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.065	0.003	0.000	0.002	1	100	0.206	0.000	10.042	10.042	See	See
8	KNeighborsClassifier_brute_force	predict	100000	1	100	0.023	0.000	0.000	0.023	1	100	0.009	0.000	2.593	2.594	See	See
10	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.765	0.021	0.000	0.003	-1	100	0.208	0.004	13.296	13.299	See	See
11	KNeighborsClassifier_brute_force	predict	100000	1	100	0.024	0.003	0.000	0.024	-1	100	0.009	0.001	2.695	2.705	See	See
13	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.047	0.002	0.000	0.002	1	5	0.168	0.000	12.177	12.177	See	See
14	KNeighborsClassifier_brute_force	predict	100000	1	100	0.023	0.001	0.000	0.023	1	5	0.009	0.000	2.608	2.608	See	See
16	KNeighborsClassifier_brute_force	predict	100000	1000	100	1.221	0.042	0.001	0.001	1	1	0.168	0.001	7.282	7.282	See	See
17	KNeighborsClassifier_brute_force	predict	100000	1	100	0.022	0.000	0.000	0.022	1	1	0.009	0.000	2.508	2.509	See	See
19	KNeighborsClassifier_brute_force	predict	100000	1000	2	1.726	0.025	0.000	0.002	-1	1	0.025	0.000	67.745	67.746	See	See
20	KNeighborsClassifier_brute_force	predict	100000	1	2	0.004	0.000	0.000	0.004	-1	1	0.001	0.000	5.696	5.704	See	See
22	KNeighborsClassifier_brute_force	predict	100000	1000	2	2.610	0.025	0.000	0.003	-1	5	0.027	0.000	98.127	98.138	See	See
23	KNeighborsClassifier_brute_force	predict	100000	1	2	0.006	0.002	0.000	0.006	-1	5	0.001	0.000	9.829	9.850	See	See
25	KNeighborsClassifier_brute_force	predict	100000	1000	2	1.971	0.003	0.000	0.002	1	100	0.060	0.002	33.026	33.038	See	See
26	KNeighborsClassifier_brute_force	predict	100000	1	2	0.003	0.000	0.000	0.003	1	100	0.001	0.000	4.089	4.103	See	See
28	KNeighborsClassifier_brute_force	predict	100000	1000	2	2.637	0.022	0.000	0.003	-1	100	0.060	0.002	44.181	44.195	See	See
29	KNeighborsClassifier_brute_force	predict	100000	1	2	0.006	0.004	0.000	0.006	-1	100	0.001	0.000	8.465	8.499	See	See
31	KNeighborsClassifier_brute_force	predict	100000	1000	2	1.947	0.001	0.000	0.002	1	5	0.027	0.000	73.332	73.334	See	See
32	KNeighborsClassifier_brute_force	predict	100000	1	2	0.003	0.000	0.000	0.003	1	5	0.001	0.000	4.479	4.483	See	See
34	KNeighborsClassifier_brute_force	predict	100000	1000	2	1.082	0.008	0.000	0.001	1	1	0.026	0.000	42.322	42.323	See	See
35	KNeighborsClassifier_brute_force	predict	100000	1	2	0.002	0.000	0.000	0.002	1	1	0.001	0.000	2.750	2.759	See	See

KNeighborsClassifier_kd_tree

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: algorithm=kd_tree.

WARNING! Mismatch between validation scores for 12 prediction (50.0%). The observed differences can be found in the diff_accuracy_scores column. The chosen difference threshold is 0.001 for accuracy_score. See details in the dataframe below.

	estimator	function	n_samples_train	n_samples	n_features	parameters_digest	dataset_digest	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	algorithm	n_jobs	n_neighbors	accuracy_score_sklearn	mean_duration_sklearnex	std_duration_sklearnex	accuracy_score_sklearnex	speedup	std_speedup	diff_accuracy_scores
1	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	fe8267e25dadb302ed498e50fabf7ef4	1eb6cf1a720a225efb91df0b529b0510	0.779112	1.022175	NaN	0.000103	0.000779	kd_tree	-1	1	0.929	0.110197	0.002212	0.910	7.070186	7.071610	0.019
4	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	dc74b969bf622bc24ba3bc62c980983b	1eb6cf1a720a225efb91df0b529b0510	0.989423	0.360718	NaN	0.000081	0.000989	kd_tree	-1	5	0.946	0.192088	0.003080	0.941	5.150893	5.151556	0.005
7	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	41accfcdcd5c50784bedf6164b63de99	1eb6cf1a720a225efb91df0b529b0510	5.436315	0.537301	NaN	0.000015	0.005436	kd_tree	1	100	0.951	0.582351	0.008237	0.940	9.335121	9.336054	0.011
10	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	022f7445d43bb1dbc24dc3106c03cb93	1eb6cf1a720a225efb91df0b529b0510	2.904067	0.263187	NaN	0.000028	0.002904	kd_tree	-1	100	0.951	0.602179	0.030179	0.940	4.822601	4.828653	0.011
13	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	84622ba45e941db642965553529e1941	1eb6cf1a720a225efb91df0b529b0510	1.618465	0.201740	NaN	0.000049	0.001618	kd_tree	1	5	0.946	0.194173	0.003210	0.941	8.335178	8.336317	0.005
16	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	a9455565db1d8e052a783317c99744ff	1eb6cf1a720a225efb91df0b529b0510	0.912997	0.326837	NaN	0.000088	0.000913	kd_tree	1	1	0.929	0.107503	0.001168	0.910	8.492755	8.493256	0.019
19	KNeighborsClassifier_kd_tree	predict	1000	1000	2	fe8267e25dadb302ed498e50fabf7ef4	edf7cb014c2c0910e1c3aaaf8ae633f2	0.024846	0.013165	NaN	0.000644	0.000025	kd_tree	-1	1	0.891	0.000405	0.000046	0.879	61.279851	61.677021	0.012
22	KNeighborsClassifier_kd_tree	predict	1000	1000	2	dc74b969bf622bc24ba3bc62c980983b	edf7cb014c2c0910e1c3aaaf8ae633f2	0.021764	0.001449	NaN	0.000735	0.000022	kd_tree	-1	5	0.911	0.000659	0.000019	0.905	33.043209	33.057182	0.006
25	KNeighborsClassifier_kd_tree	predict	1000	1000	2	41accfcdcd5c50784bedf6164b63de99	edf7cb014c2c0910e1c3aaaf8ae633f2	0.036308	0.013017	NaN	0.000441	0.000036	kd_tree	1	100	0.894	0.004508	0.000140	0.917	8.054845	8.058718	0.023
28	KNeighborsClassifier_kd_tree	predict	1000	1000	2	022f7445d43bb1dbc24dc3106c03cb93	edf7cb014c2c0910e1c3aaaf8ae633f2	0.036362	0.007232	NaN	0.000440	0.000036	kd_tree	-1	100	0.894	0.005663	0.000617	0.917	6.421526	6.459497	0.023
31	KNeighborsClassifier_kd_tree	predict	1000	1000	2	84622ba45e941db642965553529e1941	edf7cb014c2c0910e1c3aaaf8ae633f2	0.019499	0.000223	NaN	0.000821	0.000019	kd_tree	1	5	0.911	0.000649	0.000032	0.905	30.039126	30.074742	0.006
34	KNeighborsClassifier_kd_tree	predict	1000	1000	2	a9455565db1d8e052a783317c99744ff	edf7cb014c2c0910e1c3aaaf8ae633f2	0.017935	0.000134	NaN	0.000892	0.000018	kd_tree	1	1	0.891	0.000409	0.000028	0.879	43.847172	43.946977	0.012

Raw results

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	2.877	0.194	0.028	0.0	-1	1	0.764	0.128	3.768	3.821	See	See
3	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	3.551	0.162	0.023	0.0	-1	5	0.722	0.013	4.918	4.919	See	See
6	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	3.535	0.169	0.023	0.0	1	100	0.701	0.004	5.041	5.041	See	See
9	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	3.455	0.058	0.023	0.0	-1	100	0.728	0.029	4.743	4.747	See	See
12	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	3.569	0.200	0.022	0.0	1	5	0.710	0.010	5.027	5.027	See	See
15	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	3.426	0.035	0.023	0.0	1	1	0.706	0.006	4.853	4.853	See	See
18	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.001	0.022	0.0	-1	1	0.004	0.003	0.173	0.207	See	See
21	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.031	0.0	-1	5	0.002	0.002	0.329	0.520	See	See
24	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.031	0.0	1	100	0.001	0.001	0.413	0.587	See	See
27	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.031	0.0	-1	100	0.001	0.000	0.579	0.621	See	See
30	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.031	0.0	1	5	0.001	0.000	0.641	0.642	See	See
33	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.032	0.0	1	1	0.001	0.000	0.658	0.659	See	See

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.779	1.022	0.000	0.001	-1	1	0.110	0.002	7.070	7.072	See	See
2	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.002	0.000	0.000	0.002	-1	1	0.000	0.000	8.613	9.020	See	See
4	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.989	0.361	0.000	0.001	-1	5	0.192	0.003	5.151	5.152	See	See
5	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.003	0.001	0.000	0.003	-1	5	0.000	0.000	7.418	7.765	See	See
7	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	5.436	0.537	0.000	0.005	1	100	0.582	0.008	9.335	9.336	See	See
8	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.003	0.001	0.000	0.003	1	100	0.001	0.000	4.203	4.380	See	See
10	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	2.904	0.263	0.000	0.003	-1	100	0.602	0.030	4.823	4.829	See	See
11	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.005	0.001	0.000	0.005	-1	100	0.001	0.000	6.372	6.637	See	See
13	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	1.618	0.202	0.000	0.002	1	5	0.194	0.003	8.335	8.336	See	See
14	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.002	0.000	0.000	0.002	1	5	0.000	0.000	3.498	3.627	See	See
16	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.913	0.327	0.000	0.001	1	1	0.108	0.001	8.493	8.493	See	See
17	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.001	0.000	0.000	0.001	1	1	0.000	0.000	3.830	4.043	See	See
19	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.025	0.013	0.001	0.000	-1	1	0.000	0.000	61.280	61.677	See	See
20	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.002	0.000	0.000	0.002	-1	1	0.000	0.000	24.386	25.403	See	See
22	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.022	0.001	0.001	0.000	-1	5	0.001	0.000	33.043	33.057	See	See
23	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.002	0.000	0.000	0.002	-1	5	0.000	0.000	22.177	23.617	See	See
25	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.036	0.013	0.000	0.000	1	100	0.005	0.000	8.055	8.059	See	See
26	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.001	0.000	0.000	0.001	1	100	0.000	0.000	6.193	6.495	See	See
28	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.036	0.007	0.000	0.000	-1	100	0.006	0.001	6.422	6.459	See	See
29	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.002	0.000	0.000	0.002	-1	100	0.000	0.000	20.488	21.738	See	See
31	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.019	0.000	0.001	0.000	1	5	0.001	0.000	30.039	30.075	See	See
32	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.001	0.000	0.000	0.001	1	5	0.000	0.000	6.186	6.549	See	See
34	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.018	0.000	0.001	0.000	1	1	0.000	0.000	43.847	43.947	See	See
35	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.001	0.000	0.000	0.001	1	1	0.000	0.000	6.115	6.475	See	See

KMeans_tall

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: algorithm=full, n_clusters=3, max_iter=30, n_init=1, tol=1e-16.

Raw results

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	KMeans_tall	fit	1000000	1000000	2	0.515	0.085	30	0.031	0.0	random	0.370	0.027	1.390	1.393	See	See
3	KMeans_tall	fit	1000000	1000000	2	0.556	0.014	30	0.029	0.0	k-means++	0.408	0.029	1.362	1.365	See	See
6	KMeans_tall	fit	1000000	1000000	100	5.673	0.129	30	0.141	0.0	random	2.960	0.116	1.917	1.918	See	See
9	KMeans_tall	fit	1000000	1000000	100	5.868	0.015	30	0.136	0.0	k-means++	3.243	0.039	1.809	1.809	See	See

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	KMeans_tall	predict	1000000	1000	2	0.002	0.001	30	0.009	0.000	random	0.0	0.0	8.301	15.174	See	See
2	KMeans_tall	predict	1000000	1	2	0.001	0.000	30	0.000	0.001	random	0.0	0.0	8.231	13.906	See	See
4	KMeans_tall	predict	1000000	1000	2	0.002	0.001	30	0.009	0.000	k-means++	0.0	0.0	12.757	14.889	See	See
5	KMeans_tall	predict	1000000	1	2	0.001	0.000	30	0.000	0.001	k-means++	0.0	0.0	14.387	15.823	See	See
7	KMeans_tall	predict	1000000	1000	100	0.002	0.000	30	0.460	0.000	random	0.0	0.0	6.750	7.257	See	See
8	KMeans_tall	predict	1000000	1	100	0.001	0.000	30	0.001	0.001	random	0.0	0.0	12.433	13.013	See	See
10	KMeans_tall	predict	1000000	1000	100	0.002	0.000	30	0.478	0.000	k-means++	0.0	0.0	6.485	6.959	See	See
11	KMeans_tall	predict	1000000	1	100	0.001	0.000	30	0.001	0.001	k-means++	0.0	0.0	14.279	14.836	See	See

KMeans_short

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: algorithm=full, n_clusters=300, max_iter=20, n_init=1, tol=1e-16.

WARNING! Mismatch between validation scores for 4 prediction (50.0%). The observed differences can be found in the diff_adjusted_rand_scores column. The chosen difference threshold is 0.001 for adjusted_rand_score. See details in the dataframe below.

	estimator	function	n_samples_train	n_samples	n_features	parameters_digest	dataset_digest	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	algorithm	init	max_iter	n_clusters	n_init	tol	adjusted_rand_score_sklearn	mean_duration_sklearnex	std_duration_sklearnex	adjusted_rand_score_sklearnex	speedup	std_speedup	diff_adjusted_rand_scores
1	KMeans_short	predict	10000	1000	2	3db9c030f488d0e8ff350ae449da4627	058b7e3842b587a5c675518f0706f5ee	0.001707	0.000046	20	0.009373	0.000002	full	random	20	300	1	1.000000e-16	0.000126	0.000431	0.000032	-0.000965	3.957344	3.968107	0.001090
4	KMeans_short	predict	10000	1000	2	ac9b8cec4749455c0d1cd1a92c150716	058b7e3842b587a5c675518f0706f5ee	0.001893	0.000405	20	0.008454	0.000002	full	k-means++	20	300	1	1.000000e-16	0.001245	0.000421	0.000026	-0.000750	4.491481	4.500132	0.001995
7	KMeans_short	predict	10000	1000	100	3db9c030f488d0e8ff350ae449da4627	7f85b913f395ec54101a6738ea63a9a7	0.002646	0.000276	20	0.302303	0.000003	full	random	20	300	1	1.000000e-16	0.278733	0.000942	0.000063	0.293767	2.808728	2.815086	0.015034
10	KMeans_short	predict	10000	1000	100	ac9b8cec4749455c0d1cd1a92c150716	7f85b913f395ec54101a6738ea63a9a7	0.002472	0.000149	20	0.323674	0.000002	full	k-means++	20	300	1	1.000000e-16	0.317011	0.000936	0.000066	0.256968	2.639918	2.646479	0.060044

Raw results

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	KMeans_short	fit	10000	10000	2	0.073	0.000	20	0.002	0.0	random	0.026	0.004	2.771	2.797	See	See
3	KMeans_short	fit	10000	10000	2	0.212	0.002	20	0.001	0.0	k-means++	0.080	0.000	2.657	2.657	See	See
6	KMeans_short	fit	10000	10000	100	0.197	0.002	20	0.041	0.0	random	0.104	0.001	1.892	1.892	See	See
9	KMeans_short	fit	10000	10000	100	0.564	0.008	20	0.014	0.0	k-means++	0.289	0.002	1.950	1.950	See	See

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	KMeans_short	predict	10000	1000	2	0.002	0.0	20	0.009	0.000	random	0.000	0.0	3.957	3.968	See	See
2	KMeans_short	predict	10000	1	2	0.001	0.0	20	0.000	0.001	random	0.000	0.0	13.729	14.207	See	See
4	KMeans_short	predict	10000	1000	2	0.002	0.0	20	0.008	0.000	k-means++	0.000	0.0	4.491	4.500	See	See
5	KMeans_short	predict	10000	1	2	0.001	0.0	20	0.000	0.001	k-means++	0.000	0.0	12.908	13.396	See	See
7	KMeans_short	predict	10000	1000	100	0.003	0.0	20	0.302	0.000	random	0.001	0.0	2.809	2.815	See	See
8	KMeans_short	predict	10000	1	100	0.001	0.0	20	0.001	0.001	random	0.000	0.0	10.981	11.153	See	See
10	KMeans_short	predict	10000	1000	100	0.002	0.0	20	0.324	0.000	k-means++	0.001	0.0	2.640	2.646	See	See
11	KMeans_short	predict	10000	1	100	0.001	0.0	20	0.001	0.001	k-means++	0.000	0.0	10.589	10.902	See	See

LogisticRegression

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: penalty=l2, dual=False, tol=0.0001, C=1.0, fit_intercept=True, intercept_scaling=1, class_weight=nan, random_state=nan, solver=lbfgs, max_iter=100, multi_class=auto, verbose=0, warm_start=False, n_jobs=nan, l1_ratio=nan.

WARNING! Mismatch between validation scores for 2 prediction (50.0%). The observed differences can be found in the diff_accuracy_scores column. The chosen difference threshold is 0.001 for accuracy_score. See details in the dataframe below.

	estimator	function	n_samples_train	n_samples	n_features	parameters_digest	dataset_digest	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	C	class_weight	dual	fit_intercept	intercept_scaling	l1_ratio	max_iter	multi_class	n_jobs	penalty	random_state	solver	tol	verbose	warm_start	accuracy_score_sklearn	mean_duration_sklearnex	std_duration_sklearnex	accuracy_score_sklearnex	speedup	std_speedup	diff_accuracy_scores
1	LogisticRegression	predict	1000000	1000	100	f11be2529b0a77bc8fb5ee8de141e6b2	09c36ed6521e03974d0c134895f56c01	0.000412	0.000448	[20]	1.941076	4.121426e-07	1.0	NaN	False	True	1	NaN	100	auto	NaN	l2	NaN	lbfgs	0.0001	0	False	0.56	0.000864	0.001662	0.55	0.477067	1.034154	0.01
4	LogisticRegression	predict	1000	100	10000	f11be2529b0a77bc8fb5ee8de141e6b2	7daecaea01e5baa8c862f264e3c1c3b0	0.001628	0.000282	[26]	4.914008	1.627999e-05	1.0	NaN	False	True	1	NaN	100	auto	NaN	l2	NaN	lbfgs	0.0001	0	False	0.35	0.005126	0.001898	0.28	0.317605	0.338685	0.07

Raw results

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	LogisticRegression	fit	1000000	1000000	100	10.957	0.377	[20]	0.073	0.000	2.015	0.025	5.436	5.437	See	See
3	LogisticRegression	fit	1000	1000	10000	0.994	0.716	[26]	0.080	0.001	0.837	0.022	1.188	1.188	See	See

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	LogisticRegression	predict	1000000	1000	100	0.000	0.0	[20]	1.941	0.0	0.001	0.002	0.477	1.034	See	See
2	LogisticRegression	predict	1000000	1	100	0.000	0.0	[20]	0.016	0.0	0.000	0.000	0.363	0.369	See	See
4	LogisticRegression	predict	1000	100	10000	0.002	0.0	[26]	4.914	0.0	0.005	0.002	0.318	0.339	See	See
5	LogisticRegression	predict	1000	1	10000	0.000	0.0	[26]	1.066	0.0	0.002	0.000	0.047	0.047	See	See

Ridge

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: alpha=1.0, fit_intercept=True, normalize=deprecated, copy_X=True, max_iter=nan, tol=0.001, solver=auto, random_state=nan.

WARNING! Mismatch between validation scores for 1 prediction (25.0%). The observed differences can be found in the diff_r2_scores column. The chosen difference threshold is 0.001 for r2_score. See details in the dataframe below.

	estimator	function	n_samples_train	n_samples	n_features	parameters_digest	dataset_digest	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	alpha	copy_X	fit_intercept	max_iter	normalize	random_state	solver	tol	r2_score_sklearn	mean_duration_sklearnex	std_duration_sklearnex	r2_score_sklearnex	speedup	std_speedup	diff_r2_scores
1	Ridge	predict	1000	1000	10000	a70ae8dc5c059ec88cf27c61016aee2a	14fae294d530397144c65ba3209bf125	0.01211	0.000371	NaN	6.606066	0.000012	1.0	True	True	NaN	deprecated	NaN	auto	0.001	0.082567	0.019495	0.000725	0.122191	0.621177	0.621606	0.039624

Raw results

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	Ridge	fit	1000	1000	10000	0.181	0.003	0.442	0.0	0.185	0.00	0.977	0.977	See	See
3	Ridge	fit	1000000	1000000	100	1.158	0.077	0.691	0.0	0.311	0.28	3.725	5.018	See	See

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	Ridge	predict	1000	1000	10000	0.012	0.0	6.606	0.0	0.019	0.001	0.621	0.622	See	See
2	Ridge	predict	1000	1	10000	0.000	0.0	1.393	0.0	0.000	0.000	0.625	0.676	See	See
4	Ridge	predict	1000000	1000	100	0.000	0.0	5.143	0.0	0.000	0.000	0.566	0.745	See	See
5	Ridge	predict	1000000	1	100	0.000	0.0	0.017	0.0	0.000	0.000	0.611	0.647	See	See